Artificial intelligence: why ethics is a key issue
The development of artificial intelligence raises ethical questions. How much autonomy will these technologies leave to man? For what purposes will our data, on which their power is based, be used? How can we ensure that the decisions made by artificial intelligence are fair?
“You can’t run a new technology company without being optimistic. But that doesn’t mean you can’t be a naive techno-optimist. Innovation must go hand in hand with responsibility,” Microsoft CEO Satya Nadella said recently.
Present in many parts of our lives, professional as well as personal, artificial intelligence will take an increasingly important place in our daily lives. That is why it is necessary to define and anticipate risks to guard against them, as Accenture notes in its study “Technological Visions 2018”. On the 5 major trends structuring the company within three years, the firm identifies the “infallibility of data” as well as the need to set up “citizen AI”: “AI is much more than a program. She becomes a member of society who needs to be educated responsibly.”
Because if technology is neutral, it is men who define its uses. AI by nature has no conscience or morals: it is engineers and other masters of algorithms who develop the principles and its boundaries. “It’s up to us to ensure that in the very conception of AI we integrate values such as transparency, honesty upstream… “, develops the CEO of Microsoft, which advocates for a humanized vision and responsible use of AI, which aims to amplify human capabilities and ingenuity. In a recent document (*), the tech giant details its principles for successful human-robot complementarity: putting humans at the heart of the machine, democratizing AI by making it accessible to all, and finally controlling AI, guaranteeing the truthfulness, security and confidentiality of data and regulating algorithms to ensure that they integrate diversity and overcome its initial biases.
An AI shaped by all
Diversity is thus at the heart of the challenges of building and developing trust in AI. In order that the algorithms and intelligent solutions that are developed are not biased and do not promote discrimination, the plurality of design teams is a key principle to keep in mind. For Aurélie Jean, a French mathematician specializing in algorithms, integrating women, people from minorities or experts in the subject in question, helps to ensure a technological world for all:
“Studies have shown that in a group consisting of similar individuals, people focus on their similarities to develop an idea or product, while in a group consisting of different people, individuals focus on their differences but also on other potential differences.” (*)
In contrast to a “black box” that would see machines decide for us what is right, man remains at the heart of the process and master of the final choice: at the heart of the principle of transparency of algorithms, there is the need to understand the reasoning behind the decisions made by the robot.
In fact, the ethical stakes of AI question its uses: for what purpose is AI and should it be used? How can we ensure that these technologies are not manipulated for purposes contrary to the values they defend? It is among other things to answer these questions that the major digital companies founded the Partnership on AI consortium in 2016. Ethics, fairness, inclusion, confidentiality, interoperability….
The tech giants define “best practices” there so as not to take the risk that their tools will be used to violate human rights. While Google announced last June seven main principles to guide the development of artificial intelligence, Microsoft for its part called in July the US Congress for regulation of facial recognition. Initiatives that advocate for a virtuous and positive use of artificial intelligence, which, put at the service of health, education or the environment, proves to be a great tool for progress and emancipation.
To deepen your reflection on the ethical issues raised by the rise of artificial intelligence, go to the Palais des Congrès in Paris on November 6 and 7 where “Microsoft experiences18” will be held: two days of conferences where more than 300 leaders and experts will share their experiences of digital transformation.
(*) “Artificial Intelligence: survival guide. Understanding, reasoning and interacting differently with AI”, White Paper, Microsoft France, 2018.
Receive our latest news
Every day, the main Business Insider news